Upload
others
View
1
Download
0
Embed Size (px)
Citation preview
4/14/16
1
Software Testing
Overview
• What is software testing? • General testing criteria • Testing strategies • OO testing strategies • Debugging
N. Meng, B. Ryder 2
4/14/16
2
Software Testing
• Testing is the process of exercising a program with the specific intent of finding errors prior to delivery to the end user.
N. Meng, B. Ryder 3
What Does Testing Show?
N. Meng, B. Ryder 4
errors
requirements conformance
performance
an indication of quality
4/14/16
3
Verification and Validation
• Verification refers to tasks to ensure that software correctly implements a specific function – “Are we building the product right?”
• Validation refers to tasks to ensure that the built software is traceable to customer requirements – “Are we building the right product”?
N. Meng, B. Ryder 5
Who Tests the Software?
N. Meng, B. Ryder 6
developer independent tester
Understands the system but, will test "gently" and, is driven by "delivery"
Must learn about the system,
but, will attempt to break it and, is driven by quality
4/14/16
4
General Testing Criteria
• Interface integrity – Communication and collaboration between
components • Functional validity – Algorithm implementation
• Information content – Local or global data
• Performance
N. Meng, B. Ryder 7
Testing Strategies
N. Meng, B. Ryder 8
System engineering
Analysis modeling
Design modeling Code
Unit testing Integration testing
Validation testing
System testing
4/14/16
5
Testing Strategies
• We begin by “testing-in-the-small” and move toward “testing-in-the-large”
• For conventional software – The module is our initial focus – Integration of modules follows
• For OO Software – OO class is our initial focus – Integration of classes via communication
and collaboration follows
N. Meng, B. Ryder 9
Strategy 1: Unit Testing • Verification on the smallest unit of
software design
N. Meng, B. Ryder 10
N. Meng, B. Ryder
module to be
tested
test cases
results
software engineer
4/14/16
6
What Are Tested?
• Module interface – Information properly flows in and out
• Local data structures – Data stored temporarily maintains its
integrity
N. Meng, B. Ryder 11
What Are Tested?
• Boundary conditions – The module operates properly at
boundaries established to limit or restrict processing
• Independent paths – All statements in a module have been
executed at least once – Including error-handling paths
N. Meng, B. Ryder 12
4/14/16
7
Strategy 2: Integration Testing
• To construct tests to uncover errors associated with interfacing between units
• Two ways to integrate incrementally – Top-down integration – Bottom-up integration
N. Meng, B. Ryder 13
Top-Down Integration
N. Meng, B. Ryder 14
top module is tested with stubs
stubs are replaced one at a time, “depth first”
as new modules are integrated, some subset of tests is re-run
A
B
C
D E
F G
4/14/16
8
Bottom-Up Integration
N. Meng, B. Ryder 15
drivers are replaced one at a time, "depth first"
worker modules are grouped into builds and integrated
A
B
C
D E
F G
cluster
Regression Testing
• As a new module is added in integration testing, regression testing reruns some already executed tests to ensure that software changes do not cause problems in verified functions. – i.e, no unintended side effect is caused
N. Meng, B. Ryder 16
4/14/16
9
Smoke Testing
• An integration testing approach daily conducted to test time-critical projects – “The smoke test should exercise the entire
system from end to end. It does not have to be exhaustive, but it should be capable of exposing major problems. The smoke test should be thorough enough that if the build passes, you can assume that it is stable enough to be tested more thoroughly.”
N. Meng, B. Ryder 17
Smoke Testing
• Step 1: Software components already implemented are integrated into a “build” – A build includes all data files, libraries,
reusable modules, and engineered components that are required to implement one or more product functions.
N. Meng, B. Ryder 18
4/14/16
10
Smoke Testing
• Step 2: Tests are designed to expose errors that will keep the build from properly performing its function – The intent should be to uncover “show
stopper” errors that have the highest likelihood of throwing the software project behind schedule.
N. Meng, B. Ryder 19
Smoke Testing
• Step 3: The build is integrated with other builds, and the entire product is smoke tested daily. – The integration approach may be top down
or bottom up
N. Meng, B. Ryder 20
4/14/16
11
Why Smoke Testing?
• Integration risk is minimized – Uncover show-stoppers earlier
• The quality of end-product is improved – Early exposure of defects in design and
implementation • Error diagnosis and correction are
simplified – New parts are probably buggy
• Progress is easier to assess
N. Meng, B. Ryder 21
OO Testing Strategies
• Integration testing is mapped to – Thread-based testing • Integrates the classes required to respond to
one input or event for the system – Use-based testing • integrates the classes required to respond to
one use case – Cluster testing • integrates the classes required to demonstrate
one collaboration
N. Meng, B. Ryder 22
4/14/16
12
Validation Testing
• To check whether software functions as expected by customers
• To ensure that – All functional requirements are satisfied – All behavioral characteristics are achieved – All performance requirements are attained – Documentation is correct – Other requirements are met
N. Meng, B. Ryder 23
Alpha and Beta Testing
• Alpha testing – Acceptance tests by a representative group
of end users at the developer’s site for weeks or months
• Beta testing – “Live” applications of the system in an
environment with no developers’ presence
N. Meng, B. Ryder 24
4/14/16
13
Alpha Testing
• Why do we need alpha testing? – It is impossible for a developer to foresee
how customers will really use a program • How do people conduct alpha testing? – An informal “test drive” or a planned and
systematically executed series of tests – Users use the software in a natural setting
with the developers “looking over the shoulder”
N. Meng, B. Ryder 25
Beta Testing
• Why do we need beta testing? – To uncover errors that only end users seem
able to find • How do people conduct beta testing? – The customers use the software at end-
user sites – Customers record all problems that are
encountered and report them to developers at regular intervals
N. Meng, B. Ryder 26
4/14/16
14
System Testing
• A series of different tests to fully exercise the computer-based system – Recovery testing – Security testing – Stress testing – Performance testing – Deployment testing
• All tests verify that the system is successfully integrated to a larger system
N. Meng, B. Ryder 27
Recovery Testing
• To force the software to fail in a variety of ways and verify that recovery is properly performed – Automatic recovery • Evaluate whether initialization, check pointing
mechanisms, data recovery, and restart are correct – Manual recovery • Evaluate Mean-Time-To-Repair(MTTR) to
determine whether it is acceptable
N. Meng, B. Ryder 28
4/14/16
15
Security Testing
• To check whether the security protection mechanisms will actually protect the software from improper break through by: – acquiring passwords through externally – using hacking software – browsing/modifying sensitive data – intentionally causing system crash/errors
N. Meng, B. Ryder 29
Security Testing
• Given enough time and resources, good security testing will ultimately penetrate a system
• The goal of the system designer is to make penetration cost higher than the value of the information that will be obtained
N. Meng, B. Ryder 30
4/14/16
16
Stress Testing
• To execute a system by demanding resources in abnormal quantity, frequency, or volume – “How high can we crank this up before it
fails ?” • Design tests that generate ten interrupts per
second, when one or two is the average rate • Increase the input data rates by an order of
magnitude to see how input functions will respond • Design tests that may cause excessive hunting for
disk-resident data
N. Meng, B. Ryder 31
Performance Testing
• To test the run-time performance of software within the context of an integrated system
• It is usually coupled with stress testing • It requires both hardware and software
instrumentation to – measure resource utilization, e.g.,
processor cycles – monitor execution states
N. Meng, B. Ryder 32
4/14/16
17
Deployment (Configuration) Testing
• To ensure the software works in all different operating systems that it is to operate – Execute the software in each environment – Examine all installation procedures,
installer software, and user documentation
N. Meng, B. Ryder 33
When Should We Stop Testing?
• When detect some pre-defined number of errors – Use predictive models for estimation
• Examine number of errors found per unit of time – Decide whether to continue based on slope
of graphs • In reality -- WHEN YOU RUN OUT OF
TIME
N. Meng, B. Ryder 34
4/14/16
18
Debugging
• When a test uncovers an error, debugging is the process to diagnose the root cause and further remove the error
N. Meng, B. Ryder 35
The Debugging Process
N. Meng, B. Ryder 36
4/14/16
19
Why Is Debugging So Difficult?
N. Meng, B. Ryder 37
symptom cause
symptom and cause may be geographically separated symptom may disappear when another problem is fixed cause may be due to a combination of non-errors cause may be due to a system or compiler error cause may be due to assumptions that everyone believes symptom may be intermittent
The Art of Debugging
• Debugging tactics – Brute force/testing – Backtracking – Cause elimination
N. Meng, B. Ryder 38
4/14/16
20
Brute Force
• Record as much information as you can – Take memory dumps, collect runtime
traces, print or log program states • Pros – It can work when all other methods fail
• Cons – Waste effort and time – Too much information to be useful
N. Meng, B. Ryder 39
Backtracking
• Beginning at the site where a symptom has been uncovered, the source code is traced backward (manually) until the cause is found
• Pros – Simple, good for small programs
• Cons – As the number of source lines increases, the
number of potential backward paths may become unmanageably large
N. Meng, B. Ryder 40
4/14/16
21
Cause Elimination
• Data related to the error occurrence are organized to isolate potential causes – A “cause hypothesis” is devised and the
data are used to prove or disprove the hypothesis
– A list of all possible causes is developed and tests are conducted to eliminate each
N. Meng, B. Ryder 41